Sortases are bacterial enzymes capable of protein transacylation, using a wide variety of donor and acceptor molecules. The SrtA class of sortase recognizes an LPXTG sequence in a suitably engineered protein substrate or in a synthetic peptide sequence, which it cleaves between the Thr and Gly residues with concomitant formation of an acyl-enzyme intermediate. This intermediate is then resolved by nucleophilic attack, using synthetic primary amines or proteins suitably equipped with Gly or Ala as the N- terminus. Using sortases as tools, new protein labeling strategies will be developed that enable protein modifications not attainable by genetic means, such as the C-terminus to C-terminus fusion of two distinct polypeptides. These methods will be applied to the synthesis of recombinant proteins of immunological interest such as Class II MHC products, T cell receptor ectodomains and antibody F(ab) fragments, with a view to create labeled versions of these proteins that can be used for detection and isolation of their relevant counterstructures, or to enumerate the cells that bear receptors for them. The production of Class II MHC tetramers remains cumbersome, and the proposed methods have the potential of dramatically simplifying the production of these key diagnostic tools used to track pathogenic and protective T cell responses alike. The range of substrates that can be modified will be extended through the development of orthogonal labeling strategies that employ sortases of different specificities, either in their peptide recognition sequence or in their ability to accept certain types of nucleophile. This will be accomplished not only through site-directed mutagenesis of the Srt A enzymes of Staphylococcus aureus and Streptococcus pyogenes, but also through the use of other classes of sortases (e.g. SrtB from S. aureus or B. anthracis). Finally, we shall apply this technology to the question of flu particle biogenesis, a process that has so far defied observation in real time, but that may be visualized using the labeling strategies proposed here as a possible means to identify discrete steps that might serve as targets for intervention. The significance of the proposed studies lies in the development of methods that will enable the site-specific modification of proteins with entities that cannot be installed genetically.